Nvidia Anchors a US$500 Billion Financing Push for AI Data Centres

United States · TECHNOLOGY

Nvidia is assembling a US$500 billion-plus financing stack for artificial intelligence data centres, anchoring a circular system in which the chipmaker sells hardware, helps finance customers, and benefits from the resulting demand for its own products.

The Ohio project and the US$250 billion backstop

In late July 2026, multiple outlets reported that Nvidia was in talks to provide up to US$250 billion of financing support for an OpenAI data centre campus in southern Ohio. The site, roughly 50 miles south of Columbus, is being developed by SoftBank’s SB Energy and is described as a 10-gigawatt facility.

The total cost of the project could exceed US$500 billion once Nvidia chips and related equipment are included. The financing mechanism under discussion would see Nvidia provide a guarantee or backstop to help OpenAI lease computing capacity and support the broader project finance structure.

This is not a simple cash transfer. It is a structured financial commitment that turns Nvidia into both a supplier and a de facto financier for the largest AI infrastructure projects on American soil.

Wall Street treats AI data centres as a new asset class

On 27 July 2026, the BBC reported that Nvidia had struck deals with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to raise US$500 billion for artificial intelligence infrastructure. The announcement signalled that major Wall Street institutions now view graphics processing units, data centres, and long-duration compute contracts as core industrial assets.

This shift matters because it moves AI funding beyond speculative venture capital and onto the balance sheets of the world’s largest asset managers. Infrastructure funds, private credit, and sovereign capital are all flowing into a sector that barely existed at this scale five years ago.

The circular nature of the arrangement has drawn attention. Nvidia sells the chips, helps finance the customers who buy them, and then sees demand for its next-generation hardware rise as those customers expand.

The Stargate Project and the broader buildout

The Ohio discussions sit inside a wider push that began in January 2025, when OpenAI announced the Stargate Project, a joint venture with SoftBank, Oracle, and MGX. The group pledged up to US$500 billion for AI infrastructure in the United States.

In April 2025, Reuters reported that Nvidia planned to build AI servers worth as much as US$500 billion in the U.S. over four years, working with partners including TSMC. By July 2026, Nvidia and South Korea’s SK Group had unveiled a separate US$500 billion-plus AI initiative spanning data centres and next-generation memory.

Taken together, these announcements form a pattern. The United States is racing to lock up the land, power, chips, and long-term compute capacity needed to dominate the next phase of artificial intelligence.

Power, permitting, and the great-power contest

AI data centres are constrained by electricity, transmission infrastructure, and permitting timelines as much as by chip supply. A 10-gigawatt campus requires power on a scale that rivals the grid demand of entire mid-sized countries.

This makes the Ohio project geopolitically significant. The United States is trying to anchor the AI supply chain domestically, reducing dependence on foreign semiconductor fabrication and energy inputs. Control of advanced chips and the data centres that house them is now a central front in the technology rivalry with China.

Allied capital is also flowing in. Japan’s SoftBank plays a central role through SB Energy, while South Korean memory giant SK Group has its own US$500 billion-plus initiative with Nvidia. These cross-border capital flows show how allied nations are underwriting strategic infrastructure in exchange for access and influence, a dynamic explored in our pillar on Africa: The New Scramble.

What the financing model reveals about AI’s next phase

The structure of these deals points to a fundamental shift in how AI infrastructure is funded. Rather than relying solely on corporate balance sheets, the industry is building a multi-layered financing stack that includes private credit, infrastructure funds, sovereign capital, and strategic guarantees from suppliers like Nvidia.

This model raises questions about concentration risk. When the dominant chip supplier is also the dominant financier, the line between vendor and creditor blurs. Several outlets have asked whether the boom is partly self-reinforcing, with Nvidia’s financing commitments driving demand for Nvidia’s own hardware.

For now, the money is flowing. The question is whether the power grids, the permitting offices, and the physical construction timelines can keep pace with the scale of the financial commitments being made.

What to watch in the months ahead

The Ohio project remains under discussion, and no final agreement has been announced. The scale of the proposed backstop, US$250 billion, is large enough that any formal signing would be a market-moving event.

Beyond Ohio, the broader US$500 billion Wall Street financing partnership will face its first tests as projects move from term sheets to construction. Grid interconnection queues, local permitting battles, and equipment lead times will determine whether the capital can be deployed as quickly as it has been raised.

For readers tracking the intersection of technology, finance, and great-power strategy, the Nvidia-led buildout is the single most important infrastructure story of the decade. The scramble for compute is reshaping alliances, capital flows, and the physical landscape of the American Midwest.

Frequently Asked Questions

Sources

• forbes.com
• finance.yahoo.com
• reuters.com
• openai.com

This article was produced by The Rio Times’ automated newsroom system. How we use AI · Report an error

LatAm Markets: Live Signals → — real-time movers, turnover leaders and FX across Latin America.